Specific Frequencies For Gold

[ivconic]

Would be good to elaborate direct relation between tc and frequency in sampling process. Gold with its tc is a good example. It is logical to presume that Gold will be easilly detected on higher frequencies.
Another question - phase shift. Again we will take Gold to examine. Phase shift range at gold on various frequencies - is it same? (1kHz to 100kHz)

[Max]

Would be good to elaborate direct relation between tc and frequency in sampling process. Gold with its tc is a good example. It is logical to presume that Gold will be easilly detected on higher frequencies.
Another question - phase shift. Again we will take Gold to examine. Phase shift range at gold on various frequencies - is it same? (1kHz to 100kHz)

Would be good.... yes...

I think someone have made already... but there are no public data about.

It's simple... these kind of things require a good test bench and lot of time to be performed at reasonable well resolution.

I dubt anybody will find anything about... already available to the public.

Also I think math solution to the problems above is not easy... cause too many stuff involved and also shape of targets play a role I think.

Kind regards,
Max

[Tinkerer]

Would be good.... yes...

I think someone have made already... but there are no public data about.

It's simple... these kind of things require a good test bench and lot of time to be performed at reasonable well resolution.

I dubt anybody will find anything about... already available to the public.

Also I think math solution to the problems above is not easy... cause too many stuff involved and also shape of targets play a role I think.

Kind regards,
Max

Hi Max and Ivconic,

I agree that the word "frequency" is not the right one.
I am definitely not capable to express what is happening in mathematical formulas.
But I will try to explain. I start here but then I propose we open a new thread to continue the discussion.
I know that both of you have a lot of experience with metal detectors, so You will understand my less than perfect explanations.
A traditional PI provides only a small fraction of the information available in the response of the target.
An IB-PI can give much more information. This is what I have been working on with the TINKERERS IB-PI.
A picture is worth a thousand words, so I will show some scope shots.

The first pix shows the DD coil, sitting on some red clay bricks. On the coil is a transparent plastic box and on the box a steel lug.

The second pix shows the scope with the TX and RX and no target. The TX is about 90uS. The response appears during TX and after TX off. Observe how different the response is to the different targets.

The third pix shows the response of the steel lug.
The fourth pix shows the response of a 0.55gram silver ring.
The fifth pix shows the response of the steel ring.
The last pix shows the response of a 0.5gram gold ring.

The silver and gold rings are nearly the same weight, but, since gold has a specific gravity of about 19 and silver has about 11, the size is different.

The hight of the plastic box is about 30mm, so the targets are close to the coil to make the signal amplitude visible to the eye.

Enough here, lets move to a new thread.

Tinkerer

[moodz]

been there .. done that

a word of caution on these results .... very exciting on first viewing .... but and its a big BUT ... there is a big difference between sensing a change in coil inductances (due to coupling ) caused by near field inductances ( ... like rings for instance ) compared to sensing small magnetic fields due to decaying eddy fields in the far field at > 2R ( coil diameter ).

moodz.

[Tinkerer]

a word of caution on these results .... very exciting on first viewing .... but and its a big BUT ... there is a big difference between sensing a change in coil inductances (due to coupling ) caused by near field inductances ( ... like rings for instance ) compared to sensing small magnetic fields due to decaying eddy fields in the far field at > 2R ( coil diameter ).

moodz.

I agree with you.
These pictures are not showing realistic results. They are meant to be used to show the mechanics of multiple sampling, including sampling during TX.
In order to show variations in the response signal, visible to the naked eye, the targets need to be close to the coil. It is rather hard to show target responses without some signal processing.
The pictures show the signal at the output of the preamp, with a gain of about 2464. This would be a bit exaggerated it in the field.
As you see, the signal amplitude is close to saturating the opamp. Putting the targets any closer, this would happen.
By the way, the DD coil is a fairly standard type PI - DD coil.

I will be opening a new thread on the subject on the TINKERERS FORUM.
The name of the thread is "WHAT KICKS THE TARGET"
(some will say "AGAIN!!!!")

Tinkerer

[moodz]

I agree with you.
These pictures are not showing realistic results. They are meant to be used to show the mechanics of multiple sampling, including sampling during TX.
In order to show variations in the response signal, visible to the naked eye, the targets need to be close to the coil. It is rather hard to show target responses without some signal processing.
The pictures show the signal at the output of the preamp, with a gain of about 2464. This would be a bit exaggerated it in the field.
As you see, the signal amplitude is close to saturating the opamp. Putting the targets any closer, this would happen.
By the way, the DD coil is a fairly standard type PI - DD coil.

I will be opening a new thread on the subject on the TINKERERS FORUM.
The name of the thread is "WHAT KICKS THE TARGET"
(some will say "AGAIN!!!!")

Tinkerer

Point taken ... however I would like to see the response at grid points in the XYZ space above the coil . Good work though.

moodz.

[Tinkerer]

Point taken ... however I would like to see the response at grid points in the XYZ space above the coil . Good work though.

moodz.

You are welcome to test, experiment and try all of the TINKERERS ideas.
It is "OPEN SOURCE" everything is on the forum.
I welcome feedback ideas and help.

Tinkerer

[Esteban]

Imagine you can adjust step by step the pulses. An example can be like this (of course, maybe this circuit is not directly useful, just an example), in wich variable resistors can be used instead fix resistors. Also can be used more inverters. But if you're searching for specific frequency of gold (in this case the gold is more detectable with PI), you must adjust your circuits with real target, in this case nuggets, no objects as coin or others, considering difference in alloy, for example.

[Esteban]

Or maybe you need a kind of "labo", or a kind of instrument for your researchs. This is just an idea, addaptations are needed, but controls you can put in panel and study PI with this kind of instrument. So when you have fixed all parameters you want, make the specific circuit for each detector based on the circuit of your "labo". The output section you can ignore here.

Regards

[Max]

Or maybe you need a kind of "labo", or a kind of instrument for your researchs. This is just an idea, addaptations are needed, but controls you can put in panel and study PI with this kind of instrument. So when you have fixed all parameters you want, make the specific circuit for each detector based on the circuit of your "labo". The output section you can ignore here.

Regards

?

with all such controls to be operated manually and infinite variations you can have how the heck be sure you spot just right magic "frequency of gold" ?

No... if wanna best pulse kind etc you MUST use an automatic device... like change that pots with electronic pots and control them by computer/software.... then wire up some ADC and let computer find the maximum reading for gold target.

But... what about rx section of PI md ??? cause all this stuff require you have an rx section too before the adc thing.... and this will affect results maybe much more of tx pulse duration etc.

Then what about soil effects ???

You can test that thing on bench... but not on soil...

and what about e.g. orientation of target ??? you'll sure test it flat on target at a fixed distance ... but what about distance from coil to target ???

and what about coil data ??? any coil is the same ??? no...

you need a target coil also... its params , no ?

and then... what effects of shield in doing such tests ???

And damping ???

And noise level at front end ????

....etc etc etc

I think it's just a big waste of time....

It's better doing some calculations , having in mind a design and try to optimize... than discover anything about optimum "gold frequency"

Kind regards,
Max

[Esteban]

?

with all such controls to be operated manually and infinite variations you can have how the heck be sure you spot just right magic "frequency of gold" ?

No... if wanna best pulse kind etc you MUST use an automatic device... like change that pots with electronic pots and control them by computer/software.... then wire up some ADC and let computer find the maximum reading for gold target.

But... what about rx section of PI md ??? cause all this stuff require you have an rx section too before the adc thing.... and this will affect results maybe much more of tx pulse duration etc.

Then what about soil effects ???

You can test that thing on bench... but not on soil...

and what about e.g. orientation of target ??? you'll sure test it flat on target at a fixed distance ... but what about distance from coil to target ???

and what about coil data ??? any coil is the same ??? no...

you need a target coil also... its params , no ?

and then... what effects of shield in doing such tests ???

And damping ???

And noise level at front end ????

....etc etc etc

I think it's just a big waste of time....

It's better doing some calculations , having in mind a design and try to optimize... than discover anything about optimum "gold frequency"

Kind regards,
Max

OK, but calculations work better in paper but real components and all the coils aren't equals. But this is sure: you need a minilabo for to work in it. Gold frequency mean here a sensitive PI for small nuggets... of course small "nuggets" of iron can be detectable... PI don't scape of this fact.

This is not a schematic for final PI, can be a base for PI labo.

Regards

Esteban

[wyndham]

I have read the 2 pages of q&a and there is some thing that has occurred to me from time to time. No MD detects gold. The PI detects a signal, we interpret the signal to mean there's non ferrous "Thing" transmitting a RX signal back that means something to the circuit then make a noise in the earphones
Nothing that I have read has said how to increase the return signal strength or make the signal last longer. It appears from some of the post that gold dampens the signal as opposed to silver that lengthens the signal and that different alloys create something in between.
MY question is, does the signal get dampened or is it shifted as a light spectrum gets red shifted from objects moving away from us.
If it is dampened, is it because of the dielectric nature of gold and how can that characteristic be exploited?
If we continue to think that a better circuit will process a weak signal, there is a point of diminishing returns, a gold ceiling as it were.
If the signal is shifted, that should be possible to discover.
Now please understand that I have a hard time putting batteries in a flashlight and have it work but from my POV all that detector sees is a little blip of a signal and needs to be seen as 2 parts of the problem, 1. not enough signal and 2 why is the target not producing the signal to the max it could.
Just a blind squirrel looking for nut, Wyndham

[mikebg]

Specific Timeconstants of Target

Hi B^C,
Please write concise and clear questions because my machine translation can not translate meaningful your postings only. I can think about answers if I can understand your questions. Please send me your questions in this thread via Email, but edit them like for tabloid or machine. Here is the answer:
The spectral response of a gold nugget is formed by a series of time constants. This is valid for every conductive object. The first (fundamental) time constant T1 can be calculated (approximately only) by measuring nugget diameter (eddy diameter seen in direction of magnetic field) and conductivity of the goldl. It can be estimated more precisely by measuring the fundamental cutoff frequency, at which phase lag relative to excitating magnetic field (TX current) is -45 deg.
The second time constant T2 is 9 times smaller than the first. It forms second transfer channel having 9 times higher cutoff frequency. The third time constant is 25 times smaller than the first. The fourth one is 49 times smaller. General speaking, the series of time constants of every conductive target is inversely proportional to the squares of the natural series of odd numbers 1, 3, 5, 7 etc. That means, if T1=T, then the series of time constants is: T, T/9, T/25; T/49 .....
Respectively; the series of cutoff frequencies is F, 9F, 25F, 49F etc.
Target signal increases with frequency, since it is proportional to the second derivative of TX current. The weighting coefficient kn of n-th time constant Tn depends on the shape of the object. If the object is annular (like bracelet), all harmonic coefficients are infinitesimal. The nearly spherical shape of the nugget increases the weight of harmonic coefficients relative to fundamental coefficient k1.
Windham, you're wrong. The difference between the signals of two identical units in size but made of different metals, is only the size of the time constants and their coefficients. The gold gives smaller time constants relative to silver because it has less conductivity. A large nugget can generate signal as a small silver object.
I will draw a signal flow diagram that explains signal transfer by series of timeconstants, and will post it in section "General Electronics" thread "Targets Frequency Response".
Your Mike BG.

[WM6]

1.729MHz is given on the Net as to the Resonance Freq of gold. Where that comes from I have no Idea.

Small correction: resonate frequency of natural gold is at 1730 kHz and wavelenght 173 mm. Only by gold we can find this golden number: 173o/173.

[Max]

simple like that!